Pharmacokinetic/Pharmacodynamic modelling of Saxagliptin and its active metabolite, 5-hydroxy Saxagliptin in rats with Type 2 Diabetes Mellitus.

Background And Purposes: It is unclear whether the parent Saxagliptin (SAX) in vivo is the same as that in vitro, which is twice that of 5-hydroxy Saxagliptin (5-OH SAX). This study is to construct a Pharmacokinetic-Pharmacodynamic (PK-PD) link model to evaluate the genuine relationship between the concentration of parent SAX in vivo and the effect.

Methods: First, we established a reliable Ultra Performance Liquid Chromatography-Mass Spectrometry (UPLC-MS/MS) method and DPP-4 inhibition ratio determination method. Then, the T2DM rats were randomly divided into four groups, intravenous injection of 5-OH SAX (0.5 mg/kg) and saline group, intragastric administration of SAX (10 mg/kg) and Sodium carboxymethyl cellulose (CMC-Na) group. Plasma samples were collected at different time points for subsequent testing. Finally, we used the measured concentrations and inhibition ratios to construct a PK-PD link model for 5-OH SAX and parent SAX.

Results: A two-compartment with additive model showed the pharmacokinetic process of SAX and 5-OH SAX, the concentration-effect relationship was represented by a sigmoidal E model and sigmoidal E with E model for SAX and 5-OH SAX, respectively. Fitting parameters showed SAX was rapidly absorbed after administration (T=0.11 h, t=0.07 h), widely distributed in the body (V ≈ 20 L/kg), plasma exposure reached 3282.06 ng*h/mL, and the elimination half-life was 6.13 h. The maximum plasma dipeptidyl peptidase IV (DPP-4) inhibition ratio of parent SAX was 71.47%. According to the final fitting parameter EC, EC=0.46EC, it was believed that the inhibitory effect of 5-OH SAX was about half of the parent SAX, which is consistent with the literature.

Conclusions: The PK-PD link model of the parent SAX established in this study can predict its pharmacokinetic process in T2DM rats and the strength of the inhibitory effect of DPP-4 based on non-clinical data.